Previous news story    Next news story

A sneak peek at our forthcoming camera test scene

By dpreview staff on Oct 9, 2012 at 23:59 GMT

As part of the development of connect.dpreview.com we've created a more advanced, more detailed test scene, which will be launched on dpreview.com soon. The new scene is roughly eight times larger (in area) than the existing test, allowing us to test to even higher resolutions than before. A larger scene also increases working distances - avoiding the inconsistent performances we've seen from some recent fixed-lens cameras, and the new design effectively overcomes the depth of field limitations experienced with our current studio scene.

We're also working on a low-light mode that will give a truer impression of how the cameras will perform in poor lighting in the real world. The low-light mode will not only use lower luminance but also a less color-balanced light source - giving a clearer impression of what happens under real-world lighting, when cameras struggle to capture and represent color accurately.

The test scene can be seen in our phone reviews, over at connect.dpreview.com, but we'll be using it for many of the cameras launched at Photokina in the very near future, and working to re-test key cameras from the last year or two.

Comments

Total comments: 324
1234
miles green
By miles green (Oct 28, 2012)

Very nice! This will definitely lessen the OOF issues.

I'm sure you thought of a moire test, given the latest industry trends. Maybe put some textures at an angle? And some white veil for the wedding photographers? The bank note also showed that nicely.

Please also make the shaded area bigger, loved that one. How about a second row of sowing threads, right under the 1st (well-lit) row.

And some glass objects please (a mini-bar)...

0 upvotes
eric
By eric (Oct 21, 2012)

No more deep shadows. Too bad. it was nice to be able to see dynamic range.

This new test truly looks absurd. You should have some experts design a test for you instead of taking a picture of my moms refrigerator.

2 upvotes
OneGuy
By OneGuy (Oct 19, 2012)

Color has always been tricky. When looking at green on the scotch label I have no clue how good the green is (because I don't drink this brand). Same holds for "red" text. During comparisons, however, greens and reds are rendered quite differently by different cams.

The solution is to use colors via objects that we have standardized in our mind. The cardinal bird has a very specific red. (Red cherries will not do but on whip cream they would be great for another reason.) Yes, both colas have "standard" reds and kiwi fruit has its own typical green.

You got the idea and I am sure you can come up with better examples.

0 upvotes
Carbon111
By Carbon111 (Oct 16, 2012)

I miss the paperclips + green orb. I found that section very useful for specular highlights, high-contrast and purple fringing.

6 upvotes
keeponkeepingon
By keeponkeepingon (Oct 13, 2012)

"allowing us to test to even higher resolutions than before."

yet you are only publishing 8mp images on connect.

Are you doing the same for phoneless cameras?

If not why are the smartphones treated differently?

At the end of the day we want to take great pictures regardless of the tool and you do us no service by showing gimped test images.

0 upvotes
tommy leong
By tommy leong (Oct 13, 2012)

PERHAPS if we keep the old scene in the middle
and EXPAND the edges with new stuff

that would be the best compromise between the old
and new.

6 upvotes
unlearny
By unlearny (Oct 22, 2012)

This is a very good idea, one we will not explore further.

0 upvotes
coppit
By coppit (Oct 13, 2012)

It would be nice if you included a moving metronome in the low-light shot, even though catching it at the bottom of the swing might be tough. Lots of folks get cameras to shoot kids without flash, and the large sensor compact category seems to be heating up.

3 upvotes
givemealight
By givemealight (Oct 12, 2012)

I'm going to miss the blue Robot that used to be in the bottom right hand corner. It gave the whole thing a cheery demeanor

7 upvotes
guyfawkes
By guyfawkes (Oct 15, 2012)

And the Paul Smith watch face. These two items were the first I always looked at.

1 upvote
Biowizard
By Biowizard (Oct 12, 2012)

Lose the dodgy colour photos (we have no way of assessing how "good" they are to start with), add some of the old scene so we can compare new cameras to old, and put in some depth for DOF/Bokeh comparison.

Brian

6 upvotes
GregGory
By GregGory (Oct 12, 2012)

Still no backlit Stouffer wedge, bugger :((

Also, I'd have preferred to see more familiar elements, eg. the Paul Smith wrist watch, and the 1000 Lire note, excellent points of reference IMO.

At least more emphasis on the low light part of the picture, eg. the thread spools from the old setup would have been great..

I don't see any great leap forward with this setup, except for the "2D" layout.

3 upvotes
bdhufnagel
By bdhufnagel (Oct 12, 2012)

Lighting conditions are far too diffuse and even. Need a collimated source, too, along with shadow, more specular reflections, and a wider range of depth. Everything is so two dimensional and lit evenly it does not tell me much about real-world performance. At least toss a disco ball lit by spot-lights in there...

2 upvotes
Vitruvius
By Vitruvius (Oct 12, 2012)

To be constructive... I have found the paper clips with the lime green background and the dark glass bottles to be good comparison areas. They have specular highlights and are an indication of how well the sensor records real three dimensional objects with edges and very smooth gradations. The photo cards have white to black steps but the dark bottles have much more subtle transitions which can bring out JPEG banding issues. This is also handy if you plan to shoot green screen since it gives and indication of edge bleed in real life. The feathers are a good indication of how well you will be able to filter out a background with a person's hair.

Sorry, but a picture of printed pictures on a wall does not have this kind of depth for comarison.

Comment edited 1 minute after posting
5 upvotes
jgriep
By jgriep (Oct 12, 2012)

agreed keep more of the old i really liked the watch face. I've also found that halftone dots especially in a changing density area show differences in lenses at closeup distances

3 upvotes
Neodp
By Neodp (Oct 12, 2012)

No. Wrong. You have to keep a good portion of the old target, so to be able to compare older tests, to new ones. Else, you lose all credibility, as a fair an unbiased, comparative reviewer. Perhaps DP is not unbiased, and to closely tied to the sellers.

2 upvotes
Simon Joinson
By Simon Joinson (Oct 13, 2012)

Nice non sequitur

1 upvote
Neodp
By Neodp (Oct 13, 2012)

Err ah. Simon says, how would you know a "great new" camera, isn't just more of the same? Why stop the comparison, to used cameras? You would not, if you are more for the consumer, than the manufacturers. This means you may not then compare, to what you own now. Only the camera you own, and use, can you really know, and so compare "improvements", from a broader perspective.

No offense; as it looks like somebody worked real hard, and put a lot of thought, into the new impending target. However, it needs to combine, the old one.

Oculos aperi

Comment edited 48 seconds after posting
1 upvote
Corian Spirit
By Corian Spirit (Oct 12, 2012)

How about adding some tall protruding conical objects with numbered circular lines and a deep conical hole with numbered circular lines to check the DOF and natural shadows. Also, need some materials with different reflections, and some transparent or translucent object. Even more daring, add an enclosed OLED.

0 upvotes
Jonathan Lee
By Jonathan Lee (Oct 12, 2012)

i am impress with the effort. could there be some common thing found as part of daily life to give perspective?

0 upvotes
M Lammerse
By M Lammerse (Oct 12, 2012)

You mean like a sex doll?

2 upvotes
Camediadude
By Camediadude (Oct 16, 2012)

There is just something hillarious to me about Doraemon asking about a sex doll here ... I imagine though that he could readily pull out any number of sex dolls from within that 'fourth dimensional pocket' of his!

0 upvotes
tommy leong
By tommy leong (Oct 12, 2012)

and THANKS for all the effort at improving
the sites.

you are the NUMBER ONE site for digital photography ,
there is no one even close

keep up the good work.

0 upvotes
tommy leong
By tommy leong (Oct 12, 2012)

oh yes

you forgot to create some shadow areas in the scene where
high ISO testing would need to show how dark areas ( under-exposed )
are handled.

1 upvote
Richard Butler
By Richard Butler (Oct 12, 2012)

That's definitely something we'll address. The low light mode will mean there will be much more information about shadow, noise and noise-reduction behaviour.

1 upvote
tommy leong
By tommy leong (Oct 12, 2012)

i think more use of nature objects like feathers, plants, flowers etc
that has natural details would be better than using photographs or printed materials.

from my own test, more often than not, its the original material
than create a doubtful analysis later.

the use of corals, rocks,etc is great

5 upvotes
tkbslc
By tkbslc (Oct 11, 2012)

Dpreview has a real problem. They keep making the site better, but a big group of whiners are driving the regular users away by making the site less friendly. That leaves a higher concentration of sour-pusses on here which only exacerbates the problem. Pretty soon it is just going to be a handful of people who share their time between telling kids to stay off their lawn, and complaining about anything dpreview does.

2 upvotes
Biowizard
By Biowizard (Oct 12, 2012)

Another person unable to comprehend reasoned constructive criticism. You want spoon feeding? Get yourself a spoon, and feed.

2 upvotes
simon65
By simon65 (Oct 11, 2012)

I'm slightly puzzled at the use of another cameras photographs in a test scene.

If the test shot isn't sharp, how will readers know if that's due to a fault in the camera/lens being tested or is inherent to the photograph in the scene?

Surely its better to stick with items that everyones knows have an intrinsic and unvarying (over time) colour and sharpness?

6 upvotes
tkbslc
By tkbslc (Oct 11, 2012)

There are hundreds of items in that scene. Surely at least some of them will make a useful point of comparison to just about anyone.

The photograph may not be perfect (although doubtful a soft one would be chosen) but one can easily make comparisons between shots taken with different gear and determine which is getting a sharper image, or perhaps different color, etc.

0 upvotes
rhlpetrus
By rhlpetrus (Oct 11, 2012)

I agree, using real objects with lots of detail is better. We don't actually know how these images work in terms of skintones, people will start to see things with slight WB variations. Fabrics with good texture are good, as are objects with various colors. I really don't think this is an improvement over the present scene.

OTOH, the idea of using varying lighting, if well-done, is a good thing.

5 upvotes
M Lammerse
By M Lammerse (Oct 12, 2012)

Agreed, I never pay so much attention to sharpness testings. Color, DR, high iso noise etc are more interesting for most photographers

0 upvotes
Biowizard
By Biowizard (Oct 12, 2012)

I've already made that very point - why include photographs of uknown quality in a test shot? They could all have been made on blunt pinhole cameras for all we know, and mis-processed with the wrong chemistry. Apart from "standard" (ie commecial) colour charts, the ONLY thing worth including is real objects.

1 upvote
PicOne
By PicOne (Oct 13, 2012)

and won't these photos fade?

2 upvotes
Klaus dk
By Klaus dk (Oct 11, 2012)

FYI: The B&W genre scene at far left, just above the middle is a print of a work by the Danish artist Julius Exner. It is titled: "Peasants from Fanø who, in the painters absence, makes fun of his work." and was painted in 1878.

1 upvote
Dimitri Khoz
By Dimitri Khoz (Oct 11, 2012)

I also see that new testscene has 4:3 format.
How will it affect testing 3:2 format sensors?
It may not show soft corners/CA for lenses mounted on them.

Comment edited 19 seconds after posting
1 upvote
Richard Butler
By Richard Butler (Oct 11, 2012)

This is a 4:3 crop of the full scene. It's actually 16:9.

0 upvotes
Ivan Uskov
By Ivan Uskov (Oct 11, 2012)

Restore bottle. Please!

3 upvotes
Cane
By Cane (Oct 11, 2012)

Why are there always crayons!

1 upvote
tkbslc
By tkbslc (Oct 11, 2012)

Easy point of reference for color.

1 upvote
Michael_13
By Michael_13 (Oct 12, 2012)

... at least for people in countries with crayons. :-)

0 upvotes
tbcass
By tbcass (Oct 11, 2012)

The new camera test scene has both positives and negatives. On the positive side everything appears to be on the same focal plane which eliminates the problem of focal point differences. On the negative side there seems to be less dynamic range in the scene (Light and shadow areas). I also liked the glass bottles in the old one. Overall it may be better because the main complaint of the old scene was the difference in focal points when trying to compare cameras. No one test can do it all.

2 upvotes
Victor Engel
By Victor Engel (Oct 12, 2012)

In my view, everything being in the same focal plane is both a positive and a negative. I think it would be good to have a small number of items in a different plane so amount of blur resulting from sensor size is readily apparent. Yes, it also depends upon aperture. However, given the same f-number, the amount of blur will vary with sensor size. Seeing that in a test photo is a good thing, as long as it doesn't distract. So it should be very clear that it is in a different focus plane. This would be useful for testing 3D cameras, too.

0 upvotes
Dimitri Khoz
By Dimitri Khoz (Oct 11, 2012)

Comparing to the old, new test scene has no dark/shadow areas.
It lacks objects in the shadow or objects with sophisticated tonal/color range like bottles in the old scene.
Most subjects are very similar high contrast objects with small details.
Also it lacks glass objects which are usually prone to reflections.
New scene may not give real-life representation where some of the objects will always be in the shadow.
I thought it is important.

1 upvote
Renzokuken
By Renzokuken (Oct 11, 2012)

dpreview seems to be on a decline these days.

first critiques were aplenty for iPhone reviews, followed by the integration of dxO's very debatable database into their lens review, and now...

using photo prints in a camera test scene.

I find all these critiques/feedback as valid. I hope dpreview can take all these comments positively and do consider modifying a few decisions.

#1Photos prints fade overtime
#2camera used to capture the photo in the prints has its own version and method of rendering color, does not give accurate color impression to viewers
#3 print quality might affect camera test, a blurred line can be both be due to bad print quality or the camera used to take the test scene itself

please use lesser photo prints

6 upvotes
Richard Butler
By Richard Butler (Oct 11, 2012)

You're arguing that us trying to create (with the help of the feedback we're soliciting here), a new test scene that will be more informative and more meaningful is a sign of decline?

The idea that we'll put a lot of effort into re-testing cameras (some of which are no longer made) is an indication of stagnation?

Including additional information from external sources (I haven't seen any real debate about DxO's testing methodologies from people who show an understanding of what's they're reported), is a sign that we're slipping?

This is us putting in a lot of work to provide more and better information than we've ever been able to do before. I'm sorry you see that as decline.

4 upvotes
rhlpetrus
By rhlpetrus (Oct 11, 2012)

Richard, even though I don't like the use of photographs on the proposed scene (in particular those of people, WB may be an issue), congrats for trying to improve things. The point about focus is relevant, but if you are moving camera away from image, that means the DoF will be increased (please try to keep same f/number and FL for all cameras of same format).

Maybe a mix of what you are proposing and more real objects would be better. And I think using DxO MArk, including their sensor RAW specs tests, is a good thing. KUTGW! (And bring that D600's review asap, you are not waiting for the 6D to come, are you?)

Comment edited 4 minutes after posting
0 upvotes
Richard Butler
By Richard Butler (Oct 12, 2012)

The D600 review is in progress, but it's a big, complex camera, so there's plenty still to be done. We're not waiting for the 6D, though.

Comment edited 16 seconds after posting
0 upvotes
DStudio
By DStudio (Oct 12, 2012)

Come on people. If you're too dumb to recognize that the color in a test-scene photo isn't a direct representation of the camera's native color rendering, then you deserve to be confused.

0 upvotes
rich889
By rich889 (Oct 11, 2012)

the point of the comparison scene is to COMPARE between cameras across the board, OLD AND NEW. The most logical way to do that is to expand the current scene rather than delete it and start with a clean slate.

6 upvotes
roman1
By roman1 (Oct 11, 2012)

Good idea. Expanding the current scene by adding more real life objects to make the scene size larger to deal with DOF and better test high resolution cameras. Seems like this way new objectives will supplement a requirement of compatibility with old reviews too.

0 upvotes
Vitruvius
By Vitruvius (Oct 11, 2012)

Ya, I agree, good idea.... whoever came up with that idea must be a genius!

Comment edited 1 minute after posting
0 upvotes
Simon Joinson
By Simon Joinson (Oct 11, 2012)

we need a bigger scene. If we expand it and increase the shooting distance, the magnification changes and the comparability is lost. We will be reshooting around 80 cameras. We did think about this stuff.

5 upvotes
wwick
By wwick (Oct 11, 2012)

My comments when the previous test target was introduced in 2009:

"it's unfortunate that you didn't take this opportunity to make a clean break with the past and start and over with an attractive, coherent arrangement of objects that would not only be pleasing to look at in detail, but would put far greater demands on cameras."

"I think your new set should have been wider and shallower so that aperture was not a big factor. You should also have consulted an experienced prop stylist, still life photographer, or set designer so we would have something attractive to look at and you would have something less dreary with which to do your painstaking tests".

Three years later you have adopted all but the suggestion of making the scene more coherent and attractive. The flat lighting is unfortunate too. Two steps forward, one giant step back.

Comment edited 57 seconds after posting
5 upvotes
Gesture
By Gesture (Oct 11, 2012)

Exactly. Start by looking at the world's finest advertising photography in upscale magazines. Yes, you want some "targets," but I woud include items like an exquisite vase that is semi-translucent, musical instruments like a clarinet or alto sax, flower arrangement of real life flowers that have been professionally preserved, a watch and so on. Instead of crayons and watercolor pans, try pastels that are available in a complete tonal set, yet with constant Hue and Saturation (only value changes). Fabrics are good and the concept could be logically expanded.

4 upvotes
Vitruvius
By Vitruvius (Oct 11, 2012)

@ Simon Joinson -
"If we expand it and increase the shooting distance, the magnification changes and the comparability is lost. We will be reshooting around 80 cameras. We did think about this stuff."

You already compensate for the variaty of sensor sizes and crop factors presumably with lenses or changing the distance from camera to scene. It doesn't take a rocket scientist to know that you could expand the existing scene and shoot the same scene as before AND ALSO move back and shoot the entire new scene as well.

1 upvote
Simon Joinson
By Simon Joinson (Oct 11, 2012)

it takes us long enough to test cameras as it is without doubling our work. Sorry....

2 upvotes
MarkByland
By MarkByland (Oct 11, 2012)

Did DPR have to hear all of these cry-baby whinerisms when they switched up the last 3 test images or have people just sunk to even deeper depths of cry-babydom in the last 10 years?

0 upvotes
rich889
By rich889 (Oct 11, 2012)

the old scene is preferable because it gave a better mix of detailed actual objects (like the bottle label and Mickey) and printed objects (like the engraving lines of the Apollo). The vast majority of the new seems to be printed objects. Why not just expand the existing scene to give you more distance? That would make it a lot easier for us to compare older cameras (5 years ago) and the newest because it would retain some of the older objects.

6 upvotes
Vitruvius
By Vitruvius (Oct 11, 2012)

I agree with others that there at least needs to be glass and metal objects in the scene. A small light source would also be interesting. The new scene looks really flat.

4 upvotes
Vitruvius
By Vitruvius (Oct 11, 2012)

Why not simply expand the area of the existing scene around the perimeter and add a secondary camera position?

This sounds like a great idea in general but as people noted the primary point of the tool is to compare. So with an entire new scene it will not really be possible to do direct comparisons to last years cameras.

Expanding the existing scene and adding new light configurations and a secondary camera position would solve both problems during the transition time.

2 upvotes
timo
By timo (Oct 11, 2012)

Quoth R Butler (13 hours ago):
Last line of the news story: 'We'll be using it for many of the cameras launched at Photokina in the very near future, and working to re-test key cameras from the last year or two.'

Yes Richard, but most of us don't upgrade our cameras every year. It matters to me to be able to compare a new camera with one I bought FIVE years ago. And your idea of a key camera is a Canon or a Nikon. Desist. Reverse. Go back. (And chuck out that juvenile and potentially damaging voting gimmick at the same time.)

You're losing your loyal user base, chaps ...

Comment edited 27 seconds after posting
3 upvotes
Richard Butler
By Richard Butler (Oct 11, 2012)

The plan is to include more old cameras than we did last time we upgraded the test scene, so we're going to do our best to make sure we include as many relevant cameras as is sensibly possible.

These won't just be Canon and Nikon (even though, having accounted for around 80% of DSLR sales, these are the most relevant for the majority of people). We'll try to include as wide a variety and as deep a selection as we can.

1 upvote
rfsIII
By rfsIII (Oct 11, 2012)

This would be a good time to set up a schedule for replacing the objects as they fade over time—especially the ColorChecker and the other calibrated targets. It's nice to be consistent, but every dye and ink changes color as it sits.

1 upvote
DanCee
By DanCee (Oct 11, 2012)

Ok, this might be the better version, but please use it together with the old one...

5 upvotes
Matthias Hutter
By Matthias Hutter (Oct 11, 2012)

I'd like to see more:
* challenging highlights, like glass/metal reflections or maybe even a direct light source in the frame
* really dark objects, both detailed and clean (to see pattern noise)
* different material like wood, leather etc.

7 upvotes
Gesture
By Gesture (Oct 12, 2012)

Of course. I mentioned above but a smaller musical instrument like a clarinet or alto sax is just one of many detailed, exquisite real-world objects that make for better testing.

0 upvotes
timo
By timo (Oct 11, 2012)

Unnecessary. Now you won't be able to compare new cameras with the one you already have.

4 upvotes
Sten298
By Sten298 (Oct 11, 2012)

I disagree! At least you could use the old test scene together with the new one, otherwise we'll miss the possibility to compare new vs old cameras. Pls consider...

5 upvotes
de Fresz
By de Fresz (Oct 11, 2012)

I don't like it because:
– No possibility to compare newer to older cameras, scenario: I have for example D90, relatively old camera which probably wouldn't be retested and I want to compare its performance to newer ones, that's base for my buying decision.
- Lack of "3d" objects built of shadows and highlights that are the most relevant to real world photography.
- Scene is very compressed in dynamic range terms, it doesn't show at one picture performance of resolving details in shadows and highlights in accordance to ISO.
Please Dpreview, don't do it this way.

8 upvotes
increments
By increments (Oct 11, 2012)

The D90 isn't on the current studio comparison, so no need to worry.

1 upvote
Richard Butler
By Richard Butler (Oct 11, 2012)

We'll be looking to include as many older cameras as we sensibly can. We're also going to be setting it up so that the scene shows much more about DR than our current test is capable of.

We're also looking at the possibility of incorporating some more depth.

1 upvote
NektonFi
By NektonFi (Oct 11, 2012)

This is really good. And especially I liked the low light test compared to the previous version.

0 upvotes
Basco
By Basco (Oct 11, 2012)

i like it , its really bigger and better detailed ... itll show up thetiny details of the newer higher resolution beasts

Comment edited 34 seconds after posting
0 upvotes
Steve oliphant
By Steve oliphant (Oct 11, 2012)

It's great to have a target like this to compare images, but one thing that really shows noise is a red fabric with dark red pattern and light red pattern on it.This will show how nikon is very good at keeping the pattern ,but other companies like Canon and olympus you would look at the fine print and blacks .I always liked the queens face on the old target.

1 upvote
Camediadude
By Camediadude (Oct 11, 2012)

Plus one for Old Queen's face!! Oh dear, let's take a step back to assess and breathe, dpreview, and please, kindly, try to refrain from going around willy nilly throwing out baby-laden bath waters!!

2 upvotes
pieces
By pieces (Oct 11, 2012)

My 1st reference point was always the fabric swatches on the old scene. Those fabrics with the with the subtle patterns in red green and blue were great.

2 upvotes
Nightwings
By Nightwings (Oct 11, 2012)

Excellent Idea!!!!

So many camera review tests were in my opinion skewed because of potential DOF issues.

0 upvotes
JMZ48
By JMZ48 (Oct 11, 2012)

Support coments by Biowizard and Gesture below.
Also, I'd like to see 3D structures to test Front/Back focus, including micro adjustments, 3D structure to test camera/lens set for bokeh, also another 3D structure to simulate low light (night) photography to test ISO from low to highest. (similar to DCRP's SanFrancisco at night, but in lab-version so lighting will be consistant.
Left,Right,Center,Top and Bottom targets for testing focus points for focusing consistency.

2 upvotes
Dan
By Dan (Oct 11, 2012)

I would like a photo of a young Asian woman. The scene also looks kind of plain. I'm sure that a ton of thought has been put into making it technically perfect for the purpose of camera testing but it just doesn't look as pleasing as the previous scene or the one over at the Imaging Resource.

Also, will this scene adequately test shadow detail? It would also be nice if you could include elements from the old test scene like the watch, paper clips, batteries, globe, wine label, Kodak Q-60R2 target, etc. This way we can compare images from older cameras. I like the idea of the low-light mode. Maybe you could light one shot with a GE Soft White CFL and another with a plain incandescent bulb in addition to your daylight simulation.

Comment edited 4 minutes after posting
0 upvotes
Peiasdf
By Peiasdf (Oct 11, 2012)

Noooooooooo, I love the old scene/objects. I always use the battery, wine label, watch and globe to compare cameras.

BTW, without shadow, how are you going to test noise?

4 upvotes
Richard Butler
By Richard Butler (Oct 11, 2012)

With a low-light mode. The tests we're currently doing (not the version seen on Connect), suggest it will do a much better job of showing noise and noise-reduction behaviour than our current setup.

1 upvote
roman1
By roman1 (Oct 11, 2012)

Cannot really justify a reason to photograph a picture of an object rather than an actual object.
It makes me wondering how many pictures of pictures do I take?..

3 upvotes
Dan
By Dan (Oct 11, 2012)

To see how a camera renders skin tones. You could use a manikin but then you run into DOF problems again.

1 upvote
roman1
By roman1 (Oct 11, 2012)

DOF problem diminishes with distance that is increased by making the test scene larger. Exactly what they are trying to do to better test high resolution cameras.
Sure, it could be a photo to test skin tones but I see more prints than needed for this purpose.

0 upvotes
oohaah
By oohaah (Oct 10, 2012)

no, the globe is gone :(

7 upvotes
Biowizard
By Biowizard (Oct 11, 2012)

I'd blame iOS 6 and Apple Maps for that! LOL!!

Brian

2 upvotes
Gesture
By Gesture (Oct 10, 2012)

Please. Rethink the entire rationale and use of a testing scene, including taking a look at the competitors. Start with what you want to evaluate: color discrimination; fine detail; rectilinear function, uniformity of exposure across the frame, shadow detail and noise, etc. And use as many real world, 3-dimensional, non-artificially pigmented objects as possible. And an interesting, imaginative lighting scheme that really tests sensor response.

8 upvotes
Simon Joinson
By Simon Joinson (Oct 11, 2012)

What do you think our starting point was? A list of things we don't want to evaluate?

6 upvotes
Total comments: 324
1234